Abstract
Space-filling designs continue to gain popularity for computer experiments. Uniformity of space-filling characteristics has been broadly sought after to provide good estimation and prediction abilities for a variety of complex models. This article presents case studies when additional information on the features of the underlying relationship may be leveraged for selecting alternative space-filling designs that offer improvements to meet specific experimental goals. Three types of nontraditional space-filling designs are illustrated to achieve different objectives to (1) allow varied density of design points across the input space, (2) obtain balanced performance on covering the input space and the range of the response values, and (3) effectively augment existing runs to achieve certain space-filling characteristic in a sequential experiment. The mechanics for implementing these design choices are described and their flexibility to adapt to other experimental scenarios is illustrated.
Additional information
Notes on contributors
Lu Lu
Lu Lu is an Associate Professor of Statistics in the Department of Mathematics and Statistics at the University of South Florida. She was a postdoctoral research associate in the Statistical Sciences Group at Los Alamos National Laboratory. Her research interests include reliability analysis, design of experiments, response surface methodology, survey sampling, and multiple objective optimization.
Christine M. Anderson-Cook
Christine M. Anderson-Cook is a Guest Scientist in the Statistical Sciences Group at Los Alamos National Laboratory. Her research areas include reliability, design of experiments, multiple criterion optimization, and response surface methodology. She is a Fellow of the American Statistical Association and the American Society for Quality.